DESCRIPTION:The overall goals of this project are to develop accurate and realistic methods of brachytherapy dose calculation that are sufficiently fast for use in treatment planning, and to improve the accuracy with which dose distributions about Pd-103 and I-125 sources can be specified. Current dose calculation practices ignore applicator attenuation and tissue heterogeneity effects, as well as inter-source attenuation. Dose distributions around low energy sources, which are currently estimated by thermoluminescent dosimetry (TLD) measurements and/or Monte Carlo (MC) simulations, appear to have an uncertainty limited to 6-8 percent. Specific aim 1 proposes to apply rigorous numerical solutions of the fundamental Boltzmann transport equation, which are known to accurately evaluate dose in heterogeneous geometries, to the problem of calculating patient-specific 3D dose distributions. Building on past work, which demonstrates that properly optimized transport calculations are 20-fold to 200-fold faster than conventional Monte Carlo, we will continue to optimize and validate the discrete ordinates deterministic transport technique. In addition, novel variance reduction techniques will be used to accelerate forward MCPT simulations. Our goal is to develop a radiation transport simulation technique that improves upon conventional MC simulation efficiency by 100-fold, allowing full 3D dose distributions to be evaluated in less than an hour. In specific aim 2, we propose to adapt our precision radiochromic film (RCF) dosimetry system to the measurement of absolute dose about low dose-rate brachytherapy sources by improving RCF energy response, sensitivity and dependence on dose-rate. The goal is to reduce the uncertainty of brachytherapy reference dose measurements from the 7-8 percent characteristic of TLD to 3-4 percent. This should improve the empirical basis of clinical dosimetry and facilitate accurate scientific investigation of secondary charged particle transport in brachytherapy. In specific aim 3 dual-energy CT imaging will be used to define the composition, geometric architecture and dosimetric consequences of soft tissue, bone and air-cavity heterogeneities in human implant sites, with emphasis on permanent prostate implants. Specific aim 4 seeks to improve the precision and uniformity of low-energy seed dosimetry by investigating the influence of internal seed geometry on dosimetric properties and by reducing the discrepancy between measured and calculated doses.